Review of 2023 – Zoë Harcombe

Happy New Year!

We’re kicking off with a review of last year. As many people think about diet at this time of year, you may like to re-visit the June post about the DASH diet vs very low carb. You might, instead, fancy trying September’s rice diet (wink). Please don’t risk semaglutide, however, the drug covered in a February post.

The three posts in succession in July on the basics of nutrition – macronutrients, vitamins & minerals – were well received and many people have tagged these for future reference. The May post with the best rebuttals for nutritional myths was another popular one. The April post on pre-diabetes diet advice is another to bookmark, as it contains 10 basic principles of diabetes.

This was another very varied year for topics – ranging from COVID jabs & traffic crashes to gas stoves & childhood asthma. We covered red meat, carbohydrates, ultra-processed food, salt and alcohol. The conditions diabetes, cardiovascular disease, dementia and mortality were featured regularly.

A couple of big questions were also addressed. Who lives to 100? Did the Finland (North Karelia study) prove that reducing (saturated) fat reduces heart disease? And what happens when keto women switch to ‘Eatwell’?

Here’s a quick summary of each post and a link to the full article for convenience.

January

Does not having a COVID jab increase your risk of a traffic crash? This was one of the strangest titles for a Monday note yet. A novel study claimed that people who had not had a COVID vaccination had an increased risk of a traffic crash compared with those vaccinated. There were flaws with adjustments and another interesting twist undermining the study claims.

Drink more & live longer? This study claimed that low hydration was associated with higher risk of disease, aging and death. The study compared four groups of people, but the groups were so uneven that the findings applied to only 3.5% of people studied. Further investigation revealed that the most hydrated people were the most likely to die early. The implication of the paper was drink more or age faster. The missing caveat was “but not too much.”

Gas stoves & childhood asthma in the US.  This was a modelling exercise, which meant it could only produce estimates, but it went viral nonetheless. I found multiple examples of where the researchers had chosen data to suit their agenda. The study was funded by an organisation that wants a carbon free world. There was an anti-gas/pro-electric motive from the outset.

Does exercise improve the effectiveness of COVID vaccination? A study conducted in South Africa claimed that exercise improves the effectiveness of COVID vaccination. There were many issues with the paper which undermined the findings. There was an interesting finding that the researchers didn’t spot – there appeared to be a ‘sweet spot’ for vaccination.

February

A short story about a library book. A subscriber spotted a note in a keto cook-book in his local library warning against consumption of animal protein. On his behalf I approached the library and discovered that this had not been inserted by the library. More likely by a vegan activist. I also contacted the Department for Education, as the label used their logo to try to appear official.

A story about weight loss. This note returned to looking at a drug that we reviewed in 2012 – semaglutide. The drug does elicit impressive weight loss, but the side effects are serious and experienced by many people. e.g., 74% reported gastrointestinal disorders. Nearly 12 times as many people withdrew from the drug group for gastrointestinal reasons as did in the placebo group. The most serious issue was the risk of thyroid cancer. People are so desperate to lose weight that they do unwise and often dangerous things. Semaglutide strikes me as more unwise and more dangerous than most other options.

The impact of meal frequency, size & timing on weight. A study examined the association between weight change and a number of aspects of eating, which were recorded on an app. It received much media coverage. The data were unreliable – snack data were captured but excluded, minimal app usage was allowed, weight changes were self-reported and tiny.

Plant based diets & bowel cancer. This study looked at the association between plant-based diets and the risk of colorectal cancer. There were 2 findings (out of 120 possible results). Both only applied to men. The paper needed to explain why there were findings for men, not women. It didn’t. It also needed to explain that what it mostly found was no findings. It didn’t.

DASH vs American diet. The original DASH (Dietary Approaches to Stop Hypertension) trial compared DASH, a diet high in fruits and vegetables and the standard American diet for eight weeks. This study revisited the trial to look at 10-years risk of cardiovascular disease. The study essentially found that the standard American diet is bad, rather than DASH or fruit/veg diet good.

March

DIAMOND diet trial for type 2 diabetes. This note reviewed a trial protocol that plans to try to put type 2 diabetes into remission. The trial is called Dietary Approaches to the Management Of type 2 Diabetes (T2D), which has been called DIAMOND. We know already that very low carb (Virta) and very low-calorie (DiRECT – meal replacement diets) can achieve T2D remission. This trial plans to impose carb and calorie restrictions on dieters. I give my view on this.

Does Erythritol sweetener cause cardiac incidents? This study about the sweetener erythritol and cardiac events received much publicity. Erythritol is used to sweeten many drinks and processed foods. It is also used in keto foods and recipes (which attracted bashing in the media coverage). The media coverage missed the fact that this study was about blood levels of erythritol, not dietary intake of erythritol. This changed everything.

Diets, health & carbon footprint. A study evaluated six so-called popular diets for how healthy they are and for their carbon footprint. The six diets were vegan, vegetarian, pescatarian, paleo, keto and omnivore. Fewer than 1% of people studied consumed the Paleo and/or Keto diets. They were hardly popular! They were damned by modelling assumptions of course.

Carbs & cardiovascular disease. This paper was based on the UK Biobank population study. It looked for associations between types of carbohydrates (all carbs, refined, wholegrain, fibre etc.) and cardiovascular disease (all CVD, heart disease and strokes). 61 out of 63 results found nothing. There were two findings – both related to free sugars – which are known to be unhealthy. The trend for papers finding nothing but claiming otherwise continues.

April

LCHF vs HCLF for athletic performance. This was a small randomised controlled trial but all participants did both the LCHF and the HCLF interventions (31 days long) and so it could achieve significant results. The study found that weight, BMI, body fat and blood glucose were all better on the LCHF diet and athletic performance didn’t drop. The big fear of switching to fat fuelling (loss of performance) may not be warranted.

The chocolate study that fooled the world. In 2015, three German reporters set about trying to do a study, which would get published in an academic journal and generate global headlines “Chocolate helps weight loss”. They achieved their goal. This review of this classic study shows how the journalists achieved their exposé and how results were claimed despite glaring inconsistencies. Most studies that I review have flaws. This one was totally fake!

Pre-diabetes diet advice. This post was inspired by two events. First, an email containing diet advice given to someone recently diagnosed with pre-diabetes. Second, the latest document from Diabetes UK reporting that there are now five million people living with diabetes in the UK. This note explained how it is clear that the advice is causing the condition. This is a note to bookmark, as it contains 10 basic principles of diabetes (covering type 1 as well as type 2).

Low-carb diets & insulin resistance.  A study was published claiming that low-carbohydrate diets are associated with insulin resistance. The study involved 120 people assessed for 7 days. The so-called low carb diet was up to 45% of the diet in the form of carbohydrate. Low carb hasn’t been studied (45% carbohydrate is high carbohydrate in the low-carb world). If it hasn’t been studied, it can’t have been found guilty of anything.

May

Weight change & dying in older adults. This study found that older people who lost 10% of their body weight or more (during a period of a few years) had 2-4 times greater risk of dying. The implication of the study was that weight loss was a cause of death. The reality is more likely that weight loss is a marker of cancer, dementia or another condition and that the condition causes death.

The best rebuttals for nutritional myths. This note takes 10 nutritional myths and aims to provide the best rebuttal for each of them. From meat being full of saturated fat to public health having our best interests at heart. All busted! (There is a link to a presentation where I went through these myths with the Nutritional Network of Ireland).

Mediterranean diet, women & heart disease. A widely reported study claimed that adherence to ‘the Mediterranean diet’ would reduce the risk of early death in women by a quarter (relative risk). The researchers didn’t study the Mediterranean diet. They studied the fictitious Mediterranean diet, which academia has constructed. I had a rant about this, which a number of subscribers enjoyed. The absolute risk was tiny, anyway.

Heart healthy diets. In 2021, the American Heart Association (AHA) issued a scientific statement setting out 10 key principles of a heart healthy diet. A paper has been published which judged 10 diets for how well they meet the 10 AHA criteria. The AHA principles assumed that animal foods were bad and plant foods were good, fat was bad, and carbs were good etc. The assessment of diets thus came to the same conclusions.

Did Finland prove that reducing (saturated) fat reduces heart disease? This was a look at a classic study – the case of North Karelia (Finland) and the claim that reducing fat, especially saturated fat, reduces heart disease. There were many health interventions introduced in North Karelia in 1972. It cannot be assumed that dietary change made any difference when smoking more than halved, for example. There was another huge factor related to World War II. The post explains.

June

Finland, diet, cholesterol & heart disease. This was part 2 of our look at North Karelia. This looks at findings from the Seven Countries study about Finland, and further original research from my PhD – to slay the heart, cholesterol, fat assumptions still being relied upon from this era.

Healthy low carb & low fat diets. More and more studies are adding in a healthy vs unhealthy dimension when looking at diets. The assumptions about what is healthy are subjective and invariably follow the ‘plants are good, animals are bad’ doctrine. As happens so often, this study condemned low carb, but their idea of low carb was up to 42% of calorie intake, which is high carb in the low-carb world.

My chat with Ben Rubin about the UK NHS health system. Ben Rubin has been a management consultant at senior level for many years and has seen the workings of vast organisations from an interesting vantage point. He is also a trustee of the Public Health Collaboration. I heard him on another podcast mention his insights into the UK NHS and I wanted to explore these further with him.

DASH vs very low carb – an RCT. The results from a 16-week randomised controlled trial (RCT) were published. The trial compared the DASH (Dietary Approach to Stop Hypertension) diet with a (genuinely) very low-carb diet in people with hypertension, prediabetes/type 2 diabetes and overweight/obesity. The very low-carb diet performed significantly better than the DASH diet for blood pressure, blood glucose levels and weight.

Saturated fat & Parkinson’s Disease. A Swedish population study reviewed the association between dietary fat intake and the incidence of Parkinson’s Disease (PD) in 41,597 people over an average of 17.6 years. The study claimed that higher saturated fat intake might be associated with an increased risk of PD, but that there were no associations with total fat, monounsaturated fat, or polyunsaturated fat. This makes little sense if one knows basic facts about fat.

July

The basics of nutrition – macronutrients. This was the first of a really popular three part series on the basics of nutrition. What is carbohydrate? protein? fat? What role do they play in the body? What is our need for them?

The basics of nutrition – vitamins. How many vitamins are there? Which are fat soluble? water soluble? How much of each do we need? What happens if we don’t get the required amounts?

The basics of nutrition – minerals. Which minerals do we need? Where do we find them? What’s the difference between a macro mineral and a trace mineral? Metallic vs non metallic elements? What happens if we don’t get what we need?

The PURE study – diet, CVD & mortality in 80 countries. This was the latest paper from the ongoing global PURE population study. It had several aims, one of which was the development of a simple healthy diet score (based on six foods). The main difference between this diet score and others is that whole-fat dairy was included (and assumed to be health positive). The PURE diet score was tested for its relevance in many different ways in this paper. It fared well.

Keto-like diet and heart disease. I broke with my normal rule of not reviewing conference presentations to address the latest attack on keto diets. As usually happens, the researchers showed that they know what a keto diet is but then didn’t study one. They damned the diet despite not having studied it.

August

Breakfast timing & risk of type 2 diabetes. This study could be summarised as approximately 100,000 people were followed for approximately 7 years and approximately 1 in 100 people developed T2D during that time. This paper would have us believe that breakfast before 8am vs after 9am was a factor in that incidence rate. It did not make the case for that claim.

Protein & Cirrhosis of the liver. This was our first review of a paper from Iran. It should not have got through peer review, but it provided some interesting learnings. It claimed higher total protein and higher dairy protein were associated with lower deaths while higher animal protein was associated with higher deaths. These were not genuine findings. I explain why not.

MIND diet & cognitive decline. This was a Randomised Controlled Trial. These are rare in nutrition, so we should review them when they are published. This RCT included 600 people – 300 assigned to a Mediterranean/DASH type diet and 300 assigned to a control group. However, they changed calorie intake in both groups, thus negating the golden rule of RCTs. There were no findings. That could have been due to study design flaws or because the diet made no difference.

Alcohol & mortality. This was a post about a huge systematic review of all literature on alcohol and mortality. It found that associations between intake and harm occur at much higher intakes than public health bodies advise. Occasional and low volume drinkers were at no greater risk of early death than non-drinkers. There was a sex difference, however. Women could consume fewer units than men before there was a suggestion of harm.

September

Ultra-processed food & health. A story about ultra-processed foods (UPFs) made headlines. The story was generated by two, short, poster presentations at the European Society of Cardiology conference (August 2023). One was from China, the other from Australia. Neither presented evidence that warranted the extensive media coverage that this received.

Lecanemab for Alzheimer’s Disease. This was about a drug, lecanemab, which has been approved in the US for early signs of Alzheimer’s Disease. The benefits were small. The safety concerns were serious. There might be a reason that no drugs have been successful in this arena despite decades of research.

Mediterranean diet, Biobank & dementia. The UK Biobank population data were used for a study to see if a so-called Mediterranean diet is associated with lower dementia risk. The researchers looked at three measures of a Mediterranean diet, which just proves that there isn’t ‘a’ Mediterranean diet. Most subgroup analyses found nothing. No plausible mechanism was offered. This was another poor population paper.

The Rice Diet. The rice diet was featured in a presentation at a low carb conference. It dates back to the 1970s. A 1975 paper featured 106 patients who had been on the diet and who had lost at least 45kg. This post looked at how a 94-97% carbohydrate diet achieved such results.

October

LDL-C & heart disease – definitive proof? This post was prompted by an email from Professor Tim Noakes, who, in turn, had received an email from a fan of the diet-heart hypothesis. A paper about the cholesterol lowering drugs, PCSK9s, seemed to suggest that outcomes were better and adverse events no worse in the lowest LDL-C group. There were several issues with this claim.

Trans people and heart disease. Sex is a fundamental characteristic in studies of health. On average, men have more heart disease and die at an earlier age than women. I wondered if medical interventions to change sex hormone levels would impact risk rates of heart disease. There were some striking and quite shocking findings.

Does salt cause metabolic syndrome? A paper called “Dietary salt intake predicts future development of metabolic syndrome in the general population” was published by a Japanese team. The word predicts implies causation, which cannot be established in epidemiological studies. There were so many other flaws, it was difficult to know where to start.

Red meat & Type 2 Diabetes. My inbox was deluged with requests to review this paper. It was another from the Harvard school of public health. It made no sense from the outset (diabetes is an issue related to glucose and meat contains no glucose). I found 14 significant issues when I went through the paper. Some are common to all population studies; some were unique to this paper.

Who lives to 100? A novel and well conducted Swedish study reviewed 12 biomarkers to see if any were different among people who lived to 100 vs those who didn’t. The result was that 10 out of the 12 were different. The differences showed up when people were 65 to 74 and so we might be able to predict super longevity using the findings of this study.

November

What happened when keto women switched to ‘Eatwell‘? This was a fabulous study. It took 10 young, slim women who had been in ketosis for almost 4 years and tested several markers of ageing and metabolic health (BMI, insulin, fat mass etc). It then put them on the UK standard government diet for 21 days and tested the markers again (they all worsened). It then put them back on their original keto diet and tested the markers again (they all reverted to their previous good levels). That’s how quickly ‘Eatwell’ can make healthy women less healthy.

World Diabetes Day. World Diabetes Day is on 14th November each year. For this note, we looked back at progress over the past 10 years. In essence, there has been progress in research into dietary measures that can put T2D into remission. Official dietary advice has not progressed to reflect this.

Dr Jason Fung talks about dietary solutions for T2D. As a bumper bonus for World Diabetes Day we did two podcasts with two world leaders in this field. This one was with Dr Jason Fung. The second one was with…

Dr Ian Lake talks about living with T1D.

Modifiable risk factors. A large study looked at the associations between two outcomes and five risk factors. The two outcomes were cardiovascular disease (CVD) and death from any cause. The five risk factors were chosen because they are considered modifiable: BMI, systolic blood pressure, non-high-density lipoprotein cholesterol, smoking and diabetes. Two of the risk factors turned out to be far bigger risk factors than the others by a margin. One ‘risk factor’ turned out not to be.

Does cannabis help pain management? A systematic review (find all evidence) and meta-analysis (pool it together) was done to try to answer the question: does cannabis help pain management? The conclusion was that small improvements in some outcomes were reported, alongside small increased risks of adverse events.

December

Can brisk walking make you live longer? This was one of the best examples I’ve found of the healthy person confounder. When I see someone walking briskly, I think “there’s a healthy person.” I don’t think “walking briskly has made that person healthy.” The researchers should have thought the same.

Vegan vs omnivore diet. This was the latest RCT from the group that brought us DIETFITS and KETO-MED. This one involved 22 sets of twins – one assigned to a vegan diet, the other to an omnivore diet for eight weeks. Food was provided for the first four weeks. LDL-Cholesterol fell in the vegan group, but that was no surprise because their plant sterol intake increased. Weight fell in the vegan group, but that was no surprise because they were given a few hundred fewer calories. The results of this ‘research’ would have been known before it started.

The festive season, eating & weight. A thorough, systematic review examined 10 studies from the past 10 years related to festive holidays, eating and weight. All studies noted weight gain during festive periods unless interventions were made. The paper cited much other interesting literature on why this might happen. All jolly interesting.